DESCRIPTION: (Applicant's Abstract) Amphetamine and methamphetamine (A/M) use and consequences have increased dramatically in the past four years. In many regions of the U.S., treatment admissions to state and federally funded programs have almost tripled from 1992 to 1995. If these trends continue, the local treatment systems in many areas must be prepared for substantial increases in clients. However, despite the currently increasing problem, the long history of A/M use in some areas, increasing legislative interest, and beginning interest in developing specific treatment protocols, little information exists on the outcomes of in-place treatment or the natural history of users of the drug. In the proposed project, an initial interview will provide a broad base of information on A/M users, focusing on a sample of 400 A/M dependent users admired to treatment programs in Los Angeles County in 1995-96. This interview will cover the user's history prior to, during, and after the identifying treatment episode. A second interview will be conducted approximately two years later on this cohort, extending the post treatment admission history to a total of about 5 years. The study will provide a comprehensive database including demographics, family risk factors, AlDS-related risk behaviors, medical, social, and criminal history, lifetime history of drug use, treatment utilization and outcomes, and relapse patterns. The natural history methodology to be used allows construction of drug use patterns over time and examination of their relationship to treatment episode(s), background factors, other social services utilization, and patterns of criminal, health, and HIV risk behaviors. Methods of analysis will include descriptive measures, analysis of variance (ANOVA) and MANOVA, loglinear modeling, regression and discriminant analysis, survival methods, Markov models, and structural equation modeling. The primary focus of the research will be differential outcomes of A/M treatment in the community setting. More specific questions include: do A/M users respond differentially to residential versus outpatient treatment; are relapse rates related to gender, age, ethnicity, route of administration, or other client characteristics; are relapse/recovery patterns similar across successive treatment episodes; what is the decrease in criminal behavior during treatment and is it maintained after discharge; is there a multivariate model which can describe the complex interrelationships among user characteristics and their drug use and treatment careers; what are average costs of treatment per client day and how do these compare to client perceived benefits. Results will provide information on the current treatment in community-based programs for A/M use and its natural history as a basis for studying the predictors and consequences of its use and treatment outcomes over time, and for developing, implementing, and modifying prevention and treatment efforts for A/M users.